Difference Between a Trend and And an Angle
Join Date: Oct 2014
Posts: 723
So I have recently taken an interest in analyzing sports stats and I'm wondering how some of you more experienced bettors differentiate between gambler fallacy and a meaningful stat.
In theory, a previous game shouldn't have an impact on future performance, but NBA playoff teams are considerably better playing after a loss. I don't think it would be fallacious to bet on a team after a loss that has an extraordinary record after playoff losses.
Now my particular interest has been player props. Spurs game 2 Parker was -150 to hit under 1.5 three pointers. He had only covered that 2 times in all the games he played 16-17 regular season, and only 6 times in his best three point shooting year, 13-14. Well, sure enough, he covered. I can't help but wonder if this was not a good stat to look at. Today, I'm looking at Joe Johnson who is 6-9 in covering 1.5 three pointers after covering in the previous 2 games. Is that a meaningful stat or am I missing things?
Join Date: Sep 2013
Posts: 11,270
zig zag theory has been dead for a while
Join Date: Apr 2006
Posts: 241
The answer to this question usually lies in setting up a database where your independent variables are various measures of past performance weighted over some period of time and your dependent variable is of course what you are trying to predict. For example in your case I might take metrics such as three point attempts per game over the past X games, three point percentage over the past Y games, the defensive rating of the opposing team, the expected minutes played of the player vs. their typical average if you have a way to predict that, the list goes on and on. Statistical techniques can then tell you how to best weight the various factors as well as which factors are most predictive.
If you can execute this well great riches await across all areas of sports betting, good luck.
Join Date: Jul 2007
Posts: 1,148
Previous against the spread records are usually pretty meaningless. If some team/player has a good ATS record it means they either
1) Got lucky and its random or
2) The market undervalued this team/player in the past
Neither of this will help you predict the future. A legit angle is something like the market consistently underestimates rest days or overestimates the impact of weather or something.
